monetary policy discipline and macroeconomic performance: the case of indonesia

7/30/2019 Monetary Policy Discipline and Macroeconomic Performance: The Case of Indonesia

1/31

ADB EconomicsWorking Paper Series

Monetary Policy Disciplineand Macroeconomic Performance:

The Case of Indonesia

Arief Ramayandi and Aleli Rosario

No. 238 | December 2010


2/31


3/31

ADB Economics Working Paper Series No. 238

Monetary Policy Discipline

and Macroeconomic Perormance:

The Case o Indonesia

Arie Ramayandi and Aleli Rosario

December 2010

Arief Ramayandi is Economist and Aleli Rosario is Economics Ofcer in the Macroeconomics and Finance

Research Division, Economics and Research Department, Asian Development Bank. The authors aregrateful to Joseph Zveglich, Jr.; Harmanta; Tao Kong; Colin McKenzie; and participants in the Workshop on

Public Sector Economics during the 12th International Convention of the East Asian Economic Association

held 23 October 2010 in Seoul, Republic of Korea, for their helpful comments and suggestions. The

authors are responsible for all errors in the paper.


4/31

Asian Development Bank

6 ADB Avenue, Mandaluyong City

1550 Metro Manila, Philippines

www.adb.org/economics

2010 by Asian Development BankDecember 2010

ISSN 1655-5252

Publication Stock No. WPS102872

The views expressed in this paper

are those of the author(s) and do not

necessarily reect the views or policies

of the Asian Development Bank.

The ADB Economics Working Paper Series is a forum for stimulating discussion and

eliciting feedback on ongoing and recently completed research and policy studies

undertaken by the Asian Development Bank (ADB) staff, consultants, or resource

persons. The series deals with key economic and development problems, particularly

those facing the Asia and Pacic region; as well as conceptual, analytical, or

methodological issues relating to project/program economic analysis, and statistical data

and measurement. The series aims to enhance the knowledge on Asias development

and policy challenges; strengthen analytical rigor and quality of ADBs country partnership

strategies, and its subregional and country operations; and improve the quality and

availability of statistical data and development indicators for monitoring development

effectiveness.

The ADB Economics Working Paper Series is a quick-disseminating, informal publication

whose titles could subsequently be revised for publication as articles in professional

journals or chapters in books. The series is maintained by the Economics and Research

Department.


5/31

Contents

Abstract v

I. Background 1

II. Monetary Policy and Its Development in Indonesia 3

III. Methodology 4

IV. Developments in Monetary Policy Stance and Some Stylized Facts

about the Indonesian Economy 8

A. Credit Growth and Output Gap 10

B. Stock Index and Output Gap 11

V. Effects of Changes in Administered Prices on Ination

and a Corrected Measure for Indonesias Monetary Policy Stance 14

A. Impact of Changes in Administered Fuel Prices

on Aggregate CPI Ination 14

B. Correcting the Measure for Monetary Policy Stance 16

VI. Assessment 18

VII. Concluding Remarks 20

References 21


6/31


7/31

Abstract

Lax monetary policy in the United States has been pointed out as one of the

responsible factors behind the recent global crisis. Similar loose monetary

conditions also prevailed in many European countries before the crisis and were

argued to be among the accommodating factors behind the run-up in asset

prices that helped trigger the 2007 nancial market turmoil. Did a similar situation

also prevail in Asia? This paper provides an insight by specically looking at

developments in the conduct of monetary policy in Indonesia during the rst

decade of this century. It uses an estimated monetary policy rule to provide a

benchmark for assessing the actual conduct of the countrys monetary policy. Theanalysis suggests that a loose monetary policy stance also prevailed in Indonesia

in the run-up to the global nancial crisis. This situation helps to explain the

surge in the countrys ination and its very high growth in nancial condition from

late 2007 to 2008. The paper reiterates the need for monetary policy discipline

to safeguard the countrys economic stability, and provides lessons to improve its

macroeconomic management.


8/31


9/31

I. Background

Lax monetary policy in developed countries prior to the global nancial crisis is one of

the contributing factors that created the bubble in asset prices. Authorities were criticized

for keeping monetary policy too loose in the run-up to the crisis (see for example, Taylor

2007, 2008, and 2009; Ahrend et al. 2008), and for having a too narrow approach in

setting their policy (Cecchetti et al. 2000, White 2006); that is, by focusing mostly on

consumer price index (CPI) ination and, to some extent, the output gap to guide their

policy rate. This loose stance was considered to be justied given the relatively low

and stable ination environment at that time, leaving behind any pressure to tighten the

monetary policy position. The low interest rates and a stable economic condition created

a comfortable environment for people to systematically underestimate risks and put a low

risk premium on the nancial market.

Is a simple traditional monetary policy reaction function, like the Taylor rule, really silent

about the developing bubble that was left unchecked in the assets market? Taylor (2007

and 2008) suggested otherwise. Interest rates in the United States (US) after the dotcom

bubble in 2001 has been kept lower than suggested by the Taylor rule; a rule that does

not explicitly take into account asset prices. The fact that the suggested interest rate

based on the simple Taylor rule is higher than what was prescribed actually suggests

that the sign of bubble build-up in the assets market was already somewhat captured

in the arguments driving the rate resulting from the Taylor rule. Arguably, an increase

in the actual rate may not have been able to fully contain the growing bubble in asset

prices. However, the lower actual policy rate certainly provided a better environment

for the bubble to ourish. A higher policy rate would have denitely lessened supportive

conditions for the asset prices to have blown unchecked.

Ahrend et al. (2008) argue that the above monetary policy failures were not only unique

for the case of the US, but widespread in many European countries and Canada as

well. This trend for a too lax monetary policy suggests that, in the past 3 years, there

has been a tendency to shift monetary policy discipline away from what the Taylor rule

would have otherwise suggested. This relative lack of discipline has implications onthe announcement of much lower interest rates, which in turn sends false signals to

the economy by providing incentives that encourage risk-taking activities, which could

potentially send an economy into trouble.


10/31

A similar story for the case of Asia is not yet seen in the literature. Instead, the regions

monetary authorities have been praised for their successes in weathering the crisis well

and for their prudent policies and behavior prior to it (see, for example, Ito 2010 and ADB

2010). However, whether countries in Asia were really immune to the symptoms of lack

of discipline in delivering monetary policy is still unclear. This paper intends to look at thatissue by focusing on the case of Indonesia during the rst decade of this century.

To do so, the paper relies on an empirically estimated monetary policy reaction function

to provide the path of a counterfactual benchmark rate for assessing developments in

the actual movements of the policy rate, which represents the actual monetary policy

stance of the country. The benchmark rate here is derived based on an approximation of

Indonesias past behavior of monetary policy, which need not necessarily be the optimal

one.1 Therefore, what is assessed in this paper is not the deviation of an actual policy

rate from its optimum path, but rather whether there was a change in the way monetary

policy was conducted. In the event where a change is detected, one then needs to

determine if the change was for the better or otherwise.

In general, the ndings detected a change in the way monetary policy has been

conducted prior to the period when the global economic downturn affected the country.

Monetary policy in Indonesia is found to be more accommodative in the years prior to the

time when the impact of the global crisis really hit the economy. This change of monetary

policy discipline tends to be unproductive, with some unfavorable symptoms similar to

those reported in Taylor (2008) and Ahrend et al. (2008). Although the situation did not

get worse and was halted by the effect of the global crisis, there is no guarantee that

keeping such lax monetary policy would be benecial for the countrys economic stability

in the future.

The rest of the paper is organized as follows. Section II provides a brief description of the

development of monetary policy in Indonesia. Section III discusses the methodology used

to derive the benchmark rate. Section IV compares the benchmark with the actual policy

rate and presents some stylized facts about the macroeconomic condition in Indonesia

during the period where the comparison is being assessed. Section V discusses the

effect of an exogenous increase in the administered fuel prices on aggregate ination,

and corrects the measure of the benchmark rate. Section VI provides the assessment for

Indonesias monetary policy stance in the last few years. Section VII concludes.

1 The approximated monetary policy rule here does not necessarily guarantee a rule that is supposed to be

optimum or the economy. This is dierent with the Taylor rule or the case o the US, which is deemed to be an

optimal policy rule or the economy.

2 | ADB Economics Working Paper Series No. 238


11/31

II. Monetary Policy and Its Development in Indonesia

The primary goal of Indonesias central bank, Bank Indonesia (BI) has always been to

achieve and maintain stability in the value of its currency (rupiah). In pursuit of this goal,

BI adopted a crawling peg exchange rate regime during the pre-Asian crisis period.2 BIwas forced to abandon this regime due to severe pressures on the rupiah to signicantly

depreciate during the 1997/1998 Asian crisis. To arrest soaring ination and restore

condence in its currency, BI adopted a more exible regime within a tighter base

money targeting framework. To reach the base money target, BI relies on open market

operations through the sale of BI certicates (SBI).

Upon emerging from the Asian crisis, a major change in the conduct of monetary policy in

Indonesia was institutionalized. A new central banking law, enacted in 1999, established the

independence of BI, directed it to set an ination target every year, and directed monetary

policy to be geared toward the achievement of the ination target (Bank Indonesia 2000).

Following the implementation of the act, the operating target in conducting monetary policy

likewise shifted from base money targeting to interest rate targeting. The BI rate is used as

the policy instrument to direct monetary policy. Initially, the reference rate was the rate for

SBI (30 days), which was then changed into the overnight cash rate on July 2005.

The following gure displays developments in some key monetary indicators for Indonesia

during the different phases of the monetary policy regime. During the pre-Asian crisis

period, the exchange rate was very stable but depreciated sharply during the crisis.

Thereafter, the exchange rate uctuated with the adoption of a freer regime. However,

movements in foreign reserves held by BI suggest that some degree of intervention was

done to smooth exchange rate uctuations. Figure 1 also shows two noticeable jumps in

CPI ination at the later period that were attributable to changes in domestic administered

fuel prices in 2005 and 2008. This will be discussed further in Section V.

2 The International Monetary Fund classied Indonesia as adopting a managed oating exchange rate regime prior

to 1997. In practice, however, the rupiah exchange rate was practically pegged to the US dollar with a nearly xed

depreciation rate, which is normally announced once a year.

Monetary Policy Discipline and Macroeconomic Performance: The Case of Indonesia | 3


12/31

Figure 1: Movements o Monetary Indicators

-75

0

75

150

225

Q1 1995 Q1 97 Q1 99 Q1 01 Q1 03 Q1 05 Q1 07 Q1 09

Percent

CPI Exchange rate Foreign reserves

Crawling peg

regime

Asiancrisis

Base money targeting

framework

Formal ination

targeting

Monetarypolicy

Note: Exchange rate is in Indonesian rupiah/US dollar, average o period. A positive change corresponds to a depreciation.

Source: CEIC Data Company database, accessed 3 November 2010.

III. Methodology

The rule guiding monetary policy in Indonesia is approximated by an interest rate rule

that governs the path of the countrys policy rate. The monetary policy stance is decided

based on developments of a given set of critical information. Specically, the relevant

short-term interest rate is used as a proxy for the operating target to trace the historical

conduct of monetary policy. This type of policy rule typically assumes that policy makersrespond to deviations of ination from its target level and the output gap. This paper

considers a forward-looking assumption in approximating the policy rule, given the

practice of policy makers to base their policy decisions on expectations of future values of

those information variables.

To this extent, Clarida et al. (1998) proposed an estimable methodology to deal with

such forward-looking policy reaction function and demonstrated that their methodology

works well in evaluating the monetary policy behavior in G7 countries. Batini and Haldane

(1999a and 1999b) and de Brouwer and Gilbert (2005) nd that this forward-looking

specication performs better in evaluating monetary policy behavior relative to the

backward-looking one. For those reasons, the policy reaction functions in this study areestimated based on forward-looking assumptions, and the methodology adopted closely

follows that proposed by Clarida et al. (1998 and 2000).



13/31

The baseline specication for the policy rule takes a simple form. Within each of its

operating periods, the monetary authority is assumed to set the nominal interest target

rate i based on developments in the expected ination around its target and the output

gap.

i i E E x t t t n t t t q t = + ( ) + ( )+ + 1 2 * (1)

where i can be interpreted as the long-run equilibrium level of the nominal rate; is

the long-run ination target; x is the output gap that serves as a measure of cyclical

variable; and t is the set of information available to the monetary authority at the time

they set the target policy rate. Clarida et al. (1998) also entertain an extension to the

baseline model by allowing for a possibility for other variables (such as exchange rate,

money growth, international interest rate, etc.) to affect the target rate explicitly. That is:

i i E E x E z t t t n t t t q t t t k t = + ( ) + ( ) + ( )+ + + 1 2 3 * (2)

where z denotes other variable(s) affecting the target policy rate.

The policy reaction function outlined in equations (1) or (2) is often considered to be too

restrictive for describing the actual movement in the policy rate (Clarida et al. 2000).

It is restrictive in the sense that: (i) the functional form in both equations assumes

that the target rate will adjust immediately to developments of the variables that affect

it (regardless of the magnitude); (ii) they represent the systematic response of the

monetary authority to developments in the economy without acknowledging a possibility

of randomness in the policy action; and (iii) they assume that the monetary authority has

perfect control over the interest rate.

Abrupt and frequent changes in the policy rate could disrupt the capital market and erode

the credibility of a monetary authority. Since credibility is very important for a monetary

authority, it typically prefers to smooth the movements in interest rate. To avoid loss of

credibility from impulsive and large changes in the policy instrument, it is further assumed

that a monetary authority smoothes the interest rate by adjusting it partially to the target:

i i it i i t t

= ( ) + +

11

(3)

where it

is the actual interest rate at time t ; i

is the partial adjustment coefcient that

captures the degree of interest rate smoothing; and vt

is the error term introduced to

capture randomness in policy action, and the fact that a monetary authority does not haveperfect control over interest rate. The intuition behind such an adjustment scheme is that

the authority does not adjust the interest rate fully to its desired current target level, but

takes some linear combination between its desired target level and the past value of the

interest rate to smoothen its movement.



14/31

Substituting equation (1) into equation (3) to obtain an estimable equation for the policy

reaction function gives us the following:

i x it i i i t n i t q i t t = ( ) + ( ) + ( ) + ++ + 1 1 11 2 1 (4)

where

i

1=

1

and

t i t n t t n t t q t t q t E x E x = ( ) ( ) + ( ) { } ++ + + +1 1 2| | vvt

with Et t( ) = 0 . The later term t( ) is a linear combination of the forecast errors of

ination, the output gap, and the exogenous disturbance vt .

Once the estimable functional form is established, the next step would be to determine

a vector of instrumental variables (ut; u

t

tand orthogonal to t ) that includes the

monetary authoritys information set at the time they choose the interest rate. That is, the

elements of ut need also be uncorrelated with vt , hence Et t t( | ) u = 0 . The last condition

provides a basis for estimating the vector of unknown parameters 1 2 i i[ ]

using the generalized method of moments (GMM) with an optimal weighting matrix that

accounts for possible serial correlation int.

Ramayandi (2007) estimated the monetary policy reaction for Indonesia using quarterly

data from the rst quarter (1Q) of 1989 to the last quarter of 2004. He uses the 30 days

SBI rate to proxy for the policy target rate; annualized changes in the log of the headlineCPI for the measure of ination; H-P (Hodrick-Prescott) ltered seasonally adjusted real

GDP for the measure of output gap; and annualized changes in log of the nominal rupiah

exchange rate to the US dollar as other potential variable that affects the target policy

rate.

The results suggest that monetary policy in Indonesia has gone through a regime change

after the Asian nancial crisis of 19971998. Prior to the crisis, monetary policy was

driven primarily by movements in the exchange rate, and not by changes in the ination

or the output gap. This observation is in line with the crawling peg exchange rate regime

adopted by BI in the precrisis period.



15/31

In the post-Asian crisis sample, however, movements in the policy rate were no longer

driven by movements in the exchange rate. Instead, the policy rate was driven by

movements in both ination expectation and the output gap. This signies the shift in

monetary policy from a crawling peg exchange rate regime (prior to the 1997/1998 crisis)

to a Taylor type rule-based policy regime after the nancial crisis.

Table 1: Characterization o Monetary Policy Reaction Function in Indonesia

Parameterit

i

-2.73

(1.55)

1

1.78

(0.11)

2

1.04

(0.48)

i

0.52

(0.03)

Adjusted R2 0.85

Prob. value or J-stat 0.70

Note: Numbers in parentheses are the relevant standard errors.

Source: Ramayandi (2007).

Table 1 presents the summary of parameters that characterize only the post-Asian

nancial crisis monetary policy in Indonesia.3 The in-sample forecast for both precrisis

and postcrisis episodes of monetary policy is found to be quite convincing and has the

ability to closely trace the movements in the countrys policy rate (Figure 2). Particularly,

the post-Asian nancial crisis (20002004) t of the estimated monetary policy reaction

function suggests that the estimated rule captures the actual conduct of monetary policyin Indonesia very well. Therefore, the characterization of the rule can be considered as

the representation of the countrys monetary policy, and qualies to be used to derive the

benchmark path for the policy.

3 Since the pre-Asian crisis period (prior to 1997) was a crawling peg exchange rate regime, the parameters or that

period are not presented here. Reer instead to Ramayandi (2007).



16/31

Figure 2: Actual versus Benchmark Policy Rates (percent)

0

Q119

90Q1

91Q1

92Q1

93Q1

94Q1

95Q1

96Q1

97Q1

98Q1

99

Q120

00Q1

01Q1

02Q1

03Q1

04

20

40

60

80

Actual Policy Rate Benchmark Rate

Source: Authors calculation.

IV. Developments in Monetary Policy Stance

and Some Stylized Facts about the Indonesian

Economy

Given the ability of the estimated policy reaction function to trace the actual movementsas discussed in the previous section, the policy reaction function is used to simulate

the out-of-sample path for the benchmark policy rate as the basis for assessing the

development of monetary policy in Indonesia. Figure 3 presents the comparison between

the actual policy rate and its benchmark path derived from the estimated policy reaction

function.



17/31

Figure 3: Relative Monetary Policy Stance, 20012009 (percent)

0

5

10

15

20

25

Q1 2001 Q1 02 Q1 03 Q1 04 Q1 05 Q1 06 Q1 07 Q1 08 Q1 09

Actual Policy Rate Benchmark Rate

Episode 1 Episode 2


There are two episodes of huge deviations observed in Figure 3. In both cases, the

actual policy rate tends to be much lower than the benchmark rate. The rst episode is

observed between 2Q2005 to 3Q2006 (episode 1 henceforth),4 and the second comes in

4Q2007 to the end of 2008 (episode 2 henceforth). From 2005 onward, the path for the

actual policy rate is observed to be consistent with its benchmark only in the latter half of

2006 up to the rst three quarters of 2007.

Do the two episodes of deviation indicate a change in monetary policy discipline? Whateconomic factors inuenced the deviations in the policy rate? Movements in ination and

the output gap, the two variables that drive the benchmark rate, shed light on what was

going on (Figure 4). Notably, ination peaked in both episodes. Output gap, however,

behaved differently in the two episodes. Within episode 1, ination surged to about 17%

while there was no signicant pressure observed in the aggregate demand. In contrast,

episode 2 shows that a peak in ination was accompanied by a systematic pressure in

the aggregate demand that steadily went beyond the potential level of output starting in

4Q2006.

4 That is, the immediate period ater the sample considered in Ramayandis (2007) estimation.



18/31

The policy reaction function discussed in the previous section suggests that the policy

rate should have been increased in response to a rise in ination expectation and/

or output gap. This is exactly what was captured by the path of the benchmark rate in

Figure 3 (above). In episode 1, the benchmark rate increased solely on account of the

ination trend. On the other hand, the increase in episode 2 is driven by both the risingtrend of ination and the output gap. The fact that the actual policy rate was kept low

created a wedge between the series of actual policy rates and benchmark rates, which

may suggest a probable shift in the discipline of monetary policy.

Figure 4: Ination and Output Gap (percent)

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

0

3

6

9

12

15

18

Q1 2003 Q4 03 Q3 04 Q2 05 Q1 06 Q4 06 Q3 07 Q2 08 Q1 09 Q4 09

Source: CEIC Data Company database, accessed 26 August 2010.

What were the implications to the domestic nancial sector? Looking beyond the two

macro indicators may provide some stylized facts that provide further insights on the

situation.

A. Credit Growth and Output Gap

Interestingly, the growth in both private credit (overall credit of the private sector) and

consumption credit (credit obtained by individuals) show a similar pattern to that of

ination (Figure 5).5 In episode 1, the growth in private credit peaked in 3Q2005 without

any systematic pressure observed in the aggregate demand. Therefore, private creditgrowth may be considered as healthy for the economy since it did not widen output gap.

In episode 2, credit growth was not only at least 25% but it also shared a similar trend

with the burgeoning output gap. At its highest (3Q2008) in episode 2, credit growth was

accompanied by a peak in output gap. In Figure 3 (above), the policy rate was also

5 Overall private credit and consumption credit are highly correlated with a correlation coefcient o 0.995.



19/31

observed to have risen by 125 basis points from the start to end of episode 2. Such

increase, however, seemed to be not high enough based on the policy reaction function.

The lower policy rate episodes are associated with faster growth in domestic private

credit but with different implications on aggregate demand. High credit growth in episode1 seems to be healthy for supporting the pace of economic growth in the economy.

High credit growth in episode 2, on the other hand, does not seem to be as healthy

as it widened the output gap, which in turn indicates growing demand pressures in the

economy, which should have implied an increase in ination expectation and the possible

formation of a bubble within the economy.

Figure 5: Credit Growth and Output Gap (percent)

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

0

8

16

24

32

40

48

Q1 2003 Q4 03 Q3 04 Q2 05 Q1 06 Q4 06 Q3 07 Q2 08 Q1 09 Q4 09

Total Credit Growth (left)

Consumption Credit Growth (left)

Output Gap (right)


B. Stock Index and Output Gap

Figure 6 shows that while the stock market had grown at an average of 40% from 2003

to 2006, actual output on the average merely approximated potential output. In the periodconsistent with episode 1, it can be seen that when changes in the stock index peaked

on 2Q2005, aggregate demand was below potential. A similar but more obvious situation

is observed for 1Q2004. However, a sharp contrast to that pattern is observable in

episode 2, where excess demand was coupled with a declining trend in the stock index.

From the patterns for credit growth and ination, it would have been right to expect the



20/31

stock index to continue rising. The aberration in this case could be attributed to the rst-

round effects of the global crisis that overall eroded condence in the nancial markets,

resulting in the withdrawal of investments from stock markets.

Figure 6: Stock Index Change and Output Gap (percent)

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

-60

-30

0

30

60

90

120

Q1 2003 Q4 03 Q3 04 Q2 05 Q1 06 Q4 06 Q3 07 Q2 08 Q1 09 Q4 09

Stock Index Change (left) Output Gap (right)


For each of the information variables considered above, what accounts for the differing

patterns against output gap in the two episodes? Episode 1 straddles 2 years when the

Indonesian economy grew at a relatively stable rate (5.7% in 2005 and 5.5% in 2006;

see Figure 7). The growth in 2005 was remarkable considering that the economy had to

contend with, among others, the aftermath of the earthquake and tsunami of December

2004, and a drastic increase in the prices of domestic fuel as subsidies were removed.

Although private credit grew by more than 31% in 3Q2005, private spending was stied

by the ensuing soaring ination in the next quarter. The following year, 2006, was a year

of moderation as everything that surged or slumped in the previous year gradually eased

back to their usual levels (ADB 2006 and 2007).

In contrast to the challenges that episode 1 faced, episode 2 was coming from a relative

domestic boom situation. GDP grew at 6.3% in 2007, the highest since the Asian nancial

crisis a decade earlier. Private consumption and investment lifted the economy, aided by

low interest rates. Private credit rose continuously at a fast pace (ADB 2008). However,

pressure to the economy came from the global nancial crisis through exports and the

nancial markets. The rst wave of impact was felt in the stock market as the index

dived by more than 50% at the end of 2008. The second wave of impact was reected

when output went down about 1 percentage point from the average growth in posted in



21/31

20042008. Curiously, it seems that in this case, the nancial crisis aborted what could

have been an unchecked bubble in private credit in 2007.

Figure 7: GDP Growth (percent)

3.6

4.5

4.8

5.0

5.7

5.5

6.3

6.0

4.5

2.0

3.0

4.0

5.0

6.0

7.0

2001 2002 2003 2004 2005 2006 2007 2008 2009


In short, the policy rate path deviation in episode 1 occurred without any substantial

pressure from the demand side. On the other hand, the deviation in episode 2 came with

a strong indication of demand side pressure to the economy. Although the state of the

economy immediately prior to each episode was able to account for the different patterns

between output gap and each of the three information variables, it still does not explain

the reason for the path deviation in policy rate.

For this, it may be worthwhile to take a closer look at ination. From Figure 3 (above),

it can be seen that ination rose by more than 10 percentage points in 4Q2005 (within

episode 1) from its 2003 level. It stayed at double-digit levels for another three quarters

before it went down to around 6.7% up to the rst half of 2008. Then, ination shot up

once more to 12% in the second half of the year (within episode 2) before tapering down

to single-digit levels. By 1Q2010, ination was less than 4%.

What caused the sharp increases in ination during the two episodes? The succeeding

section examines at length two exogenous shocks to ination that were introduced by

the Indonesian government when it decided to change policy insofar as fuel prices are

concerned.



22/31

V. Efects o Changes in Administered Prices

on Ination and a Corrected Measureor IndonesiasMonetary Policy Stance

A. Impact o Changes in Administered Fuel Prices

on Aggregate CPI Ination

The previous section has argued that both episodes of deviation in the policy rate may

be due to the increase in the administered fuel prices brought about by the governments

decision to slash its subsidy. The increase in the administered prices could have pushed

up the aggregate CPI ination and drove the benchmark rate (Figure 2) up due to its

rule-based simulation. Meanwhile, the actual policy rate may rightly have been kept lower

since the central bank was not supposed to react to this administered prices-induced

ination. If that were the case, then the central bank may have been consistent withits way of managing monetary policy. Rather, the observed deviations demonstrate an

empirical illusion as a consequence of mechanically applying simulation according to the

prescribed rule, using only information from headline CPI ination without considering

the fact that monetary policy should not be responding to changes in the administered

components of the aggregate prices.

To test the above possibility, this paper identied events when changes in administered

fuel prices were introduced in Indonesia. Because of the erratic changes in international

oil prices and its implications to the countrys national budget, the government made

several decisions to reduce and reintroduce fuel subsidies for domestic consumption.

Each decision immediately changed prices for domestic fuel, hence altering CPI inationby shifting its mean.

Sharp increases in international oil prices forced the Indonesian government to reduce

its contribution to subsidized fuel prices for domestic consumption twice in 2005. The

subsidy contributions were reduced on 1 March and 1 October 2005. The move in March

increased average domestic fuel prices by about 29%, and the one in October jacked up

the average prices by about 100%. Other moves to alter the contribution for fuel subsidy

were implemented in 2008. In May 2008, the government cut the fuel subsidy further,

resulting in about 30% increase in average domestic fuel prices. On 1 December 2008,

however, the government reacted to the decline in international oil prices by re-increasing

its fuel subsidy contribution. This move had the effect of reducing domestic fuel prices byan average of about 15%.



23/31

Table2:ExogenousSh

ockstoInfation,2005and200

8(percent)

FuelComponent

TransportFare

Impacton

CPIInfation

Kerosene

Diesel

Gasoline

Total

City

Intercity

Taxi

Air

Sea

To

tal

Weightin2005

CPI

1.1

8

0.0

8

1.9

7

3.5

3

0.4

6

0.0

9

0.2

0

0.0

4

1Mar05

Increaseinprice

0.0

0

27.2

7

29.9

9

19.9

6

11.3

5

1.0

2

-0.2

5

0.8

3

Contributionto

CPIinfation

0.0

0

0.0

2

0.5

9

0.6

1

0.7

0

0.0

5

0.0

0

-0.0

0

0.0

0

0.7

6

1.3

7

1Oct05

Increaseinprice

115.5

1

104.7

9

79.4

5

46.3

8

35.9

0

34.0

2

5.2

0

9.6

9

Contributionto

CPIinfation

1.3

6

0.0

8

1.5

7

3.0

1

1.6

4

0.1

7

0.0

3

0.0

1

0.0

0

1.8

5

4.8

6

Weightin2008

CPI

2.6

4

0.0

8

3.7

1

2.8

5

0.6

7

0.0

7

0.4

6

0.0

0

1May08

Increaseinprice

25.0

0

27.9

1

33.3

3

23.8

6

14.9

3

14.2

9

8.7

0

0.0

0

Contributionto

CPIinfation

0.6

6

0.0

2

1.2

4

1.9

2

0.6

8

0.1

0

0.0

1

0.0

4

0.0

0

0.8

3

2.7

5

1Dec08

Increaseinprice

0.0

0

-12.7

3

-16.6

7

n.a

n.a

n.a

n.a

n.a

Contributionto

CPIinfation

0.0

0

-0.0

0

-0.2

1

-0.2

1

n.a

n.a

n.a

n.a

n.a

n

.a

-0.2

1

CPI=consumerpriceindex.

Source:AdaptedromE

conomicReportonIndonesia(BankIndonesia2005,7

7;and2008,

26).



24/31

The exogenous average price changes discussed above are only the immediate (rst

round) impact of changes in the domestic fuel subsidy on aggregate CPI ination. On

top of these, there were also second-round impacts through immediate price increases in

the transport sector. These second-round impacts, though, applied only to events when

the prices of domestic fuel increased, but not the other way around due to the downwardstickiness nature of prices. Table 2 summarizes the immediate impact of exogenous

changes in the administered fuel prices on aggregate domestic ination in Indonesia.

B. Correcting the Measure or Monetary Policy Stance

To better assess Indonesias monetary policy stance in 2005 and 2008, it is sensible

to use the above information to adjust the headline CPI ination gures to remove the

irrelevant component coming from the increase in administered fuel prices. The impact

on CPI ination gures in the last column of Table 2 provides the magnitude of mean-

shifter in headline ination gures.6 Figure 8shows that without the increased effect fromthe administered fuel prices, Indonesias CPI ination in 20052006 and in 2008 shouldhave been substantially lower than its reported headline. These adjusted ination gures

are what the central bank needs to use to when setting up the monetary policy stance.

Reacting to illusionary ination signals brought about by increases in the administered

prices, which are not in the domain of monetary policy control, would be a mistake. It

would send the economy into an overly contracted monetary condition that could result in

an unnecessary recession.

Figure 8: Adjusted CPI Ination Figures

2

6

10

14

18

Q1 2003 Q4 03 Q3 04 Q2 05 Q1 06 Q4 06 Q3 07 Q2 08 Q1 09 Q4 09

Percent,

YearonY

ear

Ination Adjusted Ination


6 Each mean-shiter stays or our quarters in this case since the year-on-year ination gure with quarterly

requency is used.



25/31

Under the assumption that the central bank was fully aware about the impact of changes

in administered fuel prices on ination, it can be expected that the banks actual monetary

policy stance would not be fully reective of developments in headline CPI ination when

changes in the administered prices take place. Factoring in the adjusted ination series

as relevant information for simulating the monetary policy stance should correct thesimulated benchmark if the central bank did not change the way it conducted monetary

policy.

Figure 9: Corrected Relative Monetary Policy Stance, 20012009 (percent)

0

4

8

12

16

20

Q12001 Q102 Q103 Q104 Q105 Q106 Q107 Q108 Q109


Figure 9 replots the relative monetary policy stance after adjusting the aggregate CPI

ination measure for the effect of changes in the administered prices of fuel. When the

actual policy rate path is plotted against the path of the corrected benchmark rate, a

different pattern from that shown in Figure 2 emerges. The deviation observed in episode

1 is no longer obvious. On the contrary, although by a slightly smaller magnitude, the

deviation observed in episode 2 is still clearly observed.

Two important messages can be drawn from the above. First, the central banks actual

monetary policy stance in episode 1 was actually consistent with the way monetary

policy had been conducted in the past. The central bank seems to have done very wellin factoring in the impact of the exogenous increase in domestic fuel prices on ination

and reacted in a timely fashion according to the appropriate signals of inationary

pressures that are controllable within the domain of its policy, without changing the way

of managing the policy itself. Second, despite the adjustment to the aggregate ination

gure, the deviation in the actual stance with respect to its benchmark path is still obvious



26/31

in episode 2. This suggests that the conduct of monetary policy may have been changed

during the episode. The central bank no longer stuck to its past discipline of conducting

monetary policy but adopted a more accommodative stance than usual.

VI. Assessment

Ination targeting as a framework for conducting monetary policy in Indonesia is still

relatively in its infancy. The framework was adopted based on National Act No. 23/1999

and its amendment, Act No. 3/2004, which mandated the central bank to focus more

closely on maintaining rupiah stability. Ination targeting framework was formally adopted

in July 2005 to strengthen the performance of Indonesias monetary policy, particularly in

curbing ination without increasing output volatility.

Although the formal adoption began only in mid-2005, the move toward conducting

monetary policy based on a Taylor type of monetary policy rule has been evident in

the post-Asian nancial crisis period (Ramayandi 2007). Prior to 2005, the path of the

policy rate in Indonesia can be well presented by a Taylor type monetary policy rule

characterized by parameters in Table 1. This suggests that the central bank of Indonesia

adopted a certain rule to discipline the conduct of its monetary policy. The discipline, to

an extent, was proven useful in helping Indonesia contain its post-Asian nancial crisis

ination in terms of lowering both the level and volatility of ination (see, for example, the

discussion in ADB 2010).

Section IV provides indications that the country changed its monetary policy discipline to

a more accommodative stance around 20072008. The lax monetary policy during this

episode is indicated by the fact that the actual policy rate has been consistently lower

than its benchmark path, derived through a policy reaction function that approximated

past monetary policy behavior. For this reason, the deviation in the stance of the

countrys monetary policy could not be interpreted in a straightforward manner as a

deviation from its optimum path, but rather only indicates a change in the way monetary

policy was conducted. Consequently, at this point there is no strong case for interpreting

the change as an undesirable one, especially since the approximated monetary policy

reaction function used to derive the benchmark path may not necessarily be the optimal

policy reaction function for the country.

The optimality issue of the approximated policy reaction function used to derive the policy

benchmark rate in this paper is discussed in Ramayandi (2009), whose analysis shows

that the policy reaction function characterized by parameters presented in Table 1 is not

necessarily the optimal policy reaction function for the case of Indonesia. He argues that

the room for the countrys central bank to improve on the performance of its monetary

policy is still wide open. The direction for improvement, however, points to a stricter



27/31

monetary policy discipline rather than a more accommodative one. Nevertheless, even

with this consideration in mind, it is still not clear if one can jump to the conclusion that

the deviation in the conduct of monetary policy in 20072008 was essentially a less

preferable one.

To make a better assessment on the deviation from past monetary policy discipline,

this paper conducted a further analysis of Indonesias macroeconomic condition during

the episode of the deviation. Ahrend et al. (2008) argues that although the episodes

of lax monetary policy (the below Taylor condition) are not a necessary condition

for nancial imbalances that potentially lead to economic troubles, the condition has

generally been associated with the build-up of nancial imbalances in housing and credit

markets. Ahrends paper also shows that the below Taylor episodes, in most cases, are

associated with very strong increases in both housing sector and private credit activities.

Discussions in the previous two sections suggest that the mound in the benchmark rate

in episode 1 is simply driven by the upward surge in ination expectation due to hugeincreases in domestic administered fuel prices. When the price adjustments in fuel prices

was factored into the policy reaction function, the ination expectation was demonstrated

to be aligned with the perception of the central bank, hence eliminating the deviation

between the actual and the benchmark policy rates. Deviations in episode 2, on the other

hand, were driven by both ination expectations and a continuous build-up of positive

output gap. The impact of both phenomena provided the impetus for the gap between the

actual and the benchmark policy rates to persist even after adjusting the ination gures.

In 2007, the Indonesian economy experienced one of its strongest growths in the decade.

The relatively high growth, however, was still mainly reliant on the strong growth in

domestic consumption (see, for example, discussion in Kong and Ramayandi 2008).Accordingly, the growth was accompanied by a build-up of positive output gap that led to

increases in ination expectations. Instead of reacting to these developments, the central

bank adopted a more accommodative monetary policy stance by not appropriately raising

its policy rate. As a result, the growth of private credit including the private consumption

credit shot to the roof in that period.

This development is very much in contrast with the situation observed in 20052006.

Although high growth in private credit was also observed in these years, it did not seem

to have negatively affected the stability of the countrys economy at that time. Output gap

was not showing any systematic deviation from zero at that time, and hence did not seem

to indicate an overheating effect on the economy.

The trend in late 2007 persisted through 3Q2008, where the very high growth in private

credit remained unchecked and accompanied by a continuously widening output gap

(Figure 4 above). The condition was comfortably accommodated by a change in the

monetary policy discipline to a more lax policy stance of the central bank at that time.



28/31

As argued in Ahrend et al. (2008), this situation suggests a potential build up of nancial

imbalances in the economy. The deviation of the monetary policy stance is also in

contrast with the suggested avenue to promote improvements in the monetary policy

performance; that is, by pursuing a stricter rather than a more lax monetary policy stance

(Ramayandi 2009). Based on these assessments, the paper concludes that the change inthe way Indonesia conducted its monetary policy in 20072008 tends to move away from,

rather than close to, its optimal one. The trend was indicating signs that the economy

may have been nurturing its own homegrown bubble, which was coincidentally aborted by

the impact of the global economic downturn around 4Q2008.

VII. Concluding Remarks

The analysis in this paper suggests that Indonesia is not excluded from the list of

countries that have shown symptoms of lack of discipline in conducting monetary policy.

The tendency started to manifest during the second half of 2007 and seems to have an

unproductive impact on keeping the economy stable. Aggregate demand has continuously

widened since the shift to a looser discipline in monetary policy took place, with

private credit also growing unchecked at the same time. Both suggest that a potential

homegrown bubble may be building up.

The above situation was reversed as the impact of the global slowdown started to

weaken the economy in the late 2008. The global crisis put the seeming trend of booming

demand in the nancial sector in Indonesia to a halt. Prior to this, there seemed to

be no signicant movement from the central bank side to moderate the rising trends.

Although the potential for a bubble build-up was there, it is not very clear how it would

have developed if the global economic downturn had not occurred. Conducting a proper

counterfactual simulation to see how the trend in 20072008 could have developed in

Indonesia sans the world economic slump would be an interesting extension for the

ndings in this paper.

The motive for the central banks shift to relax monetary policy by adopting a more

accommodative stance at that time is also unclear. Nevertheless, the tendency seemed to

have been unproductive in at least two ways. First, it potentially added to the volatility in

the countrys macroeconomic condition thereby increasing the riskiness of the economy.

Second, the shift in monetary policy discipline can also be detrimental to the centralbanks credibility if prolonged. As credibility is a virtue that is difcult to earn, it may pose

a huge cost to the central bank and needs to be seriously considered.



29/31

To avoid such things from hurting the potential growth path of the economy, bringing back

more discipline in the way monetary policy is conducted is a more productive option for

Indonesia. Focusing more on stabilizing the uctuations in aggregate domestic price level

and managing its expectation would be key for promoting and maintaining the credibility

of the central bank. This, in turn, will improve the effectiveness of the countrys monetarypolicy in promoting a stable growth of the economy.

Reerences

ADB. 2006. Asian Development Outlook 2006. Hong Kong, China: Oxford University Press for the

Asian Development Bank.

______ 2007. Asian Development Outlook 2007. Hong Kong, China.

______ 2008. Asian Development Outlook 2008. Hong Kong, China.

______ 2010. Asian Development Outlook 2010. Asian Development Bank, Manila.Ahrend, R., B. Cournede, and R. Price. 2008. Monetary Policy, Market Excesses and Financial

Turmoil. OECD Economics Department Working PaperNo. 597, Organisation for Economic

Co-operation and Development, Paris.

Bank Indonesia. 2000.Annual Report 1999. Jakarta.

. 2005. 2005 Economic Report on Indonesia. Jakarta.

. 2008. 2008 Economic Report on Indonesia. Jakarta.

Batini, N., and A. Haldane. 1999a. Forward Looking Rules for Monetary Policy. In J. B. Taylor,

ed., Monetary Policy Rules. Chicago: The University of Chicago Press for the National

Bureau of Economic Research.

. 1999b. Monetary Policy Rules and Ination Forecasts. Bank of England Quarterly Bulletin

39(1):607.

Cecchetti, S. G., H. Genberg, J. Lipsky, and S. Wadhwani. 2000. Asset Prices and Central Bank

Policy. Geneva Reports on the World Economy No. 2, International Center for Monetary andBanking Studies, Geneva.

Clarida, R., J. Gali, and M. Gertler. 1998. Monetary Policy Rules in Practice: Some International

Evidence. European Economic Review42(6):103367.

. 2000. Monetary Policy Rules and Macroeconomic Stability: Evidence and Some Theory.

The Quarterly Journal of Economics 115(1):14780.

de Brouwer, G. J., and J. Gilbert. 2005. Monetary Policy Reaction Functions in Australia. The

Economic Record81(253):12434.

Ito, T. 2010. Monetary Policy and Financial Stability: Is Ination Targeting Pass? ADB Economics

Working Paper Series No. 206, Economics and Research Department, Asian Development

Bank, Manila.

Kong, T., and A. Ramayandi. 2008. Survey of Recent Development. Bulletin of Indonesian

Economic Studies 44(1):732.

Ramayandi, A. 2007. Approximating Monetary Policy: Case Study for the ASEAN-5. WorkingPapers in Economics and Development Studies 200707, Department of Economics,

Padjadjaran University. Available: www.equitablepolicy.org/wpaper/200707.pdf.

. 2009. Assessing Monetary Policy Efciency in the ASEAN-5 Countries. Working Papers

in Economics and Development Studies 200902, Department of Economics, Padjadjaran

University. Available: www.equitablepolicy.org/wpaper/200902.pdf.

http://wpfileshr2/ERD/EROD/2008%20monographs%20for%20processing/WP/EWP%20238%20-%20Aleli%20and%20Arief/www.equitablepolicy.org/wpaper/200707.pdfhttp://wpfileshr2/ERD/EROD/2008%20monographs%20for%20processing/WP/EWP%20238%20-%20Aleli%20and%20Arief/www.equitablepolicy.org/wpaper/200902.pdfhttp://wpfileshr2/ERD/EROD/2008%20monographs%20for%20processing/WP/EWP%20238%20-%20Aleli%20and%20Arief/www.equitablepolicy.org/wpaper/200902.pdfhttp://wpfileshr2/ERD/EROD/2008%20monographs%20for%20processing/WP/EWP%20238%20-%20Aleli%20and%20Arief/www.equitablepolicy.org/wpaper/200707.pdf


30/31

Taylor, J. B. 2007. Housing and Monetary Policy. Background paper for the Federal Reserve

Bank of Kansas City Annual Conference on Housing, Housing Finance, and Monetary Policy,

30 August1 September 2007, Kansas City.

. 2008. The Financial Crisis and the Policy Responses: An Empirical Analysis of What

Went Wrong. InA Festschrift in Honour of David Dodges Contributions to Canadian Public

Policy. Bank of Canada, Ottawa.. 2009. Getting Off Track: How Government Actions and Interventions Caused, Prolonged,

and Worsened the Financial Crisis. Hoover Institution, Stanford. Processed.

White, W. R. 2006. Procyclicality in the Financial System: Do We Need a New Macronancial

Stabilisation Framework? BIS Working Paper No. 193, Bank for International Settlements,

Basel.



31/31

About the Paper

Arie Ramayandi and Aleli Rosario examine the conduct o monetary discipline in Indonesia

using an estimated monetary policy rule as benchmark. Their analysis suggests that a loosemonetary policy stance had prevailed in Indonesia, which helps to explain the surges in the

countrys ination and fnancial condition rom late 2007 to 2008. They reiterate the need

or monetary policy discipline to saeguard the country s economic stability, and provide

lessons to improve its macroeconomic management.

About the Asian Development Bank

ADBs vision is an Asia and Pacifc region ree o poverty. Its mission is to help its developing

member countries substantially reduce poverty and improve the quality o lie o their

people. Despite the regions many successes, it remains home to two-thirds o the worlds

poor: 1.8 billion people who live on less than $2 a day, with 903 million struggling on

less than $1.25 a day. ADB is committed to reducing poverty through inclusive economic

growth, environmentally sustainable growth, and regional integration.Based in Manila, ADB is owned by 67 members, including 48 rom the region. Its

main instruments or helping its developing member countries are policy dialogue, loans,equity investments, guarantees, grants, and technical assistance.

Asian Development Bank

6 ADB Avenue, Mandaluyong City

1550 Metro Manila, Philippines

www.adb.org/economics

ISSN: 1655-5252

Publication Stock No. WPS102872 Printed in the Philippines

< 0 1 0 2 8 7 2 0 >

monetary policy discipline and macroeconomic performance: the case of indonesia

Documents